OBJECT-CENTRIC AND MEMORY-GUIDED NORMALITY RECONSTRUCTION FOR VIDEO ANOMALY DETECTION

Khalil Bergaoui , Yassine Naji , Aleksandr Setkov , Angelique Loesch , Michèle Gouiffès (1) , Romaric Audigier

1 AMI - Architecture et Modèles pour l'Interaction

This paper addresses the anomaly detection problem for videosurveillance. Due to the inherent rarity and heterogeneity of abnormal events, this problem is tackled from a normality modeling perspective, where our model learns object-centric normal patterns without seeing anomalous samples during training. Our main contributions consist in coupling objectlevel action features with a cosine distance-based anomaly estimation function. We therefore extend previous methods by introducing explicit geometric constraints to the mainstream reconstruction-based strategy. Our framework leverages both appearance and motion information to learn object-level behavior and captures prototypical patterns within a memory module. Experiments on several well-known datasets demonstrate the effectiveness of our method as it outperforms current state-of-the-art on most relevant spatio-temporal evaluation metrics.

Keywords: deep learning abnormal event detection video anomaly detection object-centric normality modeling; deep learning; abnormal event detection; video anomaly detection; object-centric normality modeling

Domains: Computer Science [cs]; Computer Vision and Pattern Recognition [cs.CV]

Khalil Bergaoui, Yassine Naji, Aleksandr Setkov, Angelique Loesch, Michèle Gouiffès, et al.. OBJECT-CENTRIC AND MEMORY-GUIDED NORMALITY RECONSTRUCTION FOR VIDEO ANOMALY DETECTION. 2022 IEEE International Conference on Image Processing (ICIP), Oct 2022, Bordeaux, France. pp.2691-2695, ⟨10.1109/ICIP46576.2022.9897259⟩. ⟨hal-03880897⟩